Skip to content

Conversation

@nielsbauman
Copy link
Contributor

The end time of a TSDB data stream is updated by
UpdateTimeSeriesRangeService every 5 minutes. This means that every once in a while the test fails if UpdateTimeSeriesRangeService happens to have run during the test. This PR gets rid of this check.

It's a bit odd that this failure hasn't shown up before (i.e. outside of the multi-project test suite), but I suspect it's related to the race condition we're seeing here - the MP test suite is likely a bit slower, making it slightly more likely to occur, perhaps.

Resolves #124575

The end time of a TSDB data stream is updated by
`UpdateTimeSeriesRangeService` every 5 minutes. This means that every
once in a while the test fails if UpdateTimeSeriesRangeService happens
to have run during the test. This PR gets rid of this check.

It's a bit odd that this failure hasn't shown up before (i.e. outside of
the multi-project test suite), but I suspect it's related to the race
condition we're seeing here - the MP test suite is likely a bit slower,
making it slightly more likely to occur, perhaps.

Resolves elastic#124575
@nielsbauman nielsbauman added >test Issues or PRs that are addressing/adding tests :Data Management/Data streams Data streams and their lifecycles :StorageEngine/TSDB You know, for Metrics labels Jul 7, 2025
@nielsbauman nielsbauman requested a review from masseyke July 7, 2025 19:08
@elasticsearchmachine
Copy link
Collaborator

Pinging @elastic/es-storage-engine (Team:StorageEngine)

@elasticsearchmachine
Copy link
Collaborator

Pinging @elastic/es-data-management (Team:Data Management)

Copy link
Member

@masseyke masseyke left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Seems like a reasonable theory to me. I'm also concerned about why it wasn't doing this before. Did you accidentally fix UpdateTimeSeriesRangeService while adding multiproject support?

@nielsbauman
Copy link
Contributor Author

I just realized that the cause is probably that we run these tests with a slightly different setup in the MP test suite. The data streams module has its own dedicated YAML test suite, but we're simply throwing it on a heap of YAML tests when we run the MP YAML test suite (with a slightly different cluster configuration). That will change in the future (every test suite will run in MP mode on its own, instead of running one big MP suite).

I still think it's fine to merge this fix (instead of just muting the test in MP temporarily), as it still feels like a race condition to me.

@nielsbauman nielsbauman merged commit 560464e into elastic:main Jul 8, 2025
33 checks passed
@nielsbauman nielsbauman deleted the fix-124575 branch July 8, 2025 00:36
@martijnvg
Copy link
Member

This fix makes sense to me too.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

:Data Management/Data streams Data streams and their lifecycles :StorageEngine/TSDB You know, for Metrics Team:Data Management Meta label for data/management team Team:StorageEngine >test Issues or PRs that are addressing/adding tests v9.2.0

Projects

None yet

Development

Successfully merging this pull request may close these issues.

[CI] CoreWithMultipleProjectsClientYamlTestSuiteIT test {yaml=data_stream/150_tsdb/created the data stream} failing

4 participants